yandex_api.py 文件源码

def query(self, query, num_results=30):
        query = urllib2.quote(self.preprocess(query))
        p = 0
        next_url = self.base_url.replace("{query}", query).replace("{page}", str(p))
        all_images = []
        for i in range(0, num_results, self.max_number_per_req):
            page = html.fromstring(self.opener.open(next_url).read())
            elements = page.find_class('serp-list')
            e = elements[0].getchildren()
            images = [json.loads(el.get('data-bem'))['serp-item'] for el in e]
            for image in images:
                all_images.append({'image_id': image['id'], 'url': image['img_href']})
            p += 1
            next_url = self.base_url.replace("{query}", query).replace("{page}", str(p))
            time.sleep(0.25)
        return all_images[:num_results]